From owner-freebsd-scsi Sun Mar 9 06:29:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA23717 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 06:29:10 -0800 (PST) Received: from weenix.guru.org (kmitch@weenix.guru.org [198.82.200.65]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA23712 for ; Sun, 9 Mar 1997 06:29:08 -0800 (PST) Received: (from kmitch@localhost) by weenix.guru.org (8.8.5/8.8.4) id JAA00689 for scsi@freebsd.org; Sun, 9 Mar 1997 09:29:07 -0500 (EST) From: Keith Mitchell Message-Id: <199703091429.JAA00689@weenix.guru.org> Subject: Micropolis 4221W and 20MB/s SYNC To: scsi@freebsd.org Date: Sun, 9 Mar 1997 09:29:07 -0500 (EST) X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk For some reason my Micropolis 4221W (wide scsi) drive is only getting 10 MB/s sync rate instead of the 20 it is supposed to get (according to the drive spec). The drive is hanging off an Adaptec 3940UW controller. The sync rate in the setup is set to 20 and ultra is disabled (enabling it doesn't do anything). I am running 2.2-GAMMA from 3/05. Below is the verbose probe message for the drive. ahc0: target 0 using 16Bit transfers ahc0: target 0 synchronous at 10.0MHz, offset = 0x8 ahc0: target 0 Tagged Queuing Device (ahc0:0:0): "MICROP 4221-09 1128RF 28RF" type 0 fixed SCSI 2 sd is configured at 0 sd0(ahc0:0:0): Direct-Access 1955MB (4004219 512 byte sectors) sd0(ahc0:0:0): with 4048 cyls, 9 heads, and an average 109 sectors/track As you can see it uses wide transfers. It just is operating at hakf the speed it should. From owner-freebsd-scsi Sun Mar 9 07:25:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA25073 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 07:25:15 -0800 (PST) Received: from m1.cs.man.ac.uk (0@m1.cs.man.ac.uk [130.88.13.4]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA25068 for ; Sun, 9 Mar 1997 07:25:12 -0800 (PST) Received: from amu5.cs.man.ac.uk by m1.cs.man.ac.uk (4.1/SMI-4.1:AL6) id AA10544; Sun, 9 Mar 97 15:25:04 GMT Received: from gort (annex1-8.mcc.ac.uk) by amu5.cs.man.ac.uk; Sun, 9 Mar 97 15:25:01 GMT Message-Id: <3322D5EA.9398D0B@cs.man.ac.uk> Date: Sun, 09 Mar 1997 15:23:22 +0000 From: Dave Gilbert Organization: University of Manchester Computer Science Department X-Mailer: Mozilla 3.01 (X11; I; Linux 2.1.24 i586) Mime-Version: 1.0 To: scsi@freebsd.org Subject: Medium error messages - how to read Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Our 9GB Micropolis 1990 has just been moved from a Sun where it was working fine to our PC with FreeBSD nad probably at the time the first backup dump was run we've seen the following in our logs: Mar 8 21:57:40 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb asc:11,0 U nrecovered read error Mar 8 21:57:40 uriah /kernel: , retries:4 Mar 8 21:57:40 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb asc:11,0 U nrecovered read error Mar 8 21:57:40 uriah /kernel: , retries:3 Mar 8 21:57:41 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb asc:11,0 U nrecovered read error Mar 8 21:57:41 uriah /kernel: , retries:2 Mar 8 21:57:41 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb asc:11,0 U nrecovered read error Mar 8 21:57:41 uriah /kernel: , retries:1 B Mar 8 21:57:42 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb asc:11,0 U nrecovered read error Mar 8 21:57:42 uriah /kernel: , FAILURE Now I'm presuming that the 2940UW driver is not lying and it really is a Medium error. First of all what should I do - is there anyway I can see if the drive has automatic block remapping on? Failing that how do I read the error - what partition is it on? Is there anyway I can track down the file in question? I'm inclined to think its the drive; but the fact that its only happened after its been put on the PC makes me suspicious - even though it is the 3rd one of these drives that we've had! Dave -- ----------- (Phone: 0161-275-3547) ------------------------ Man can not live - David Alan Gilbert - gilbertd@cs.man.ac.uk - G7FHJ@GB7BEV by bread alone. He ----------- (University of Manchester - AMULET Group) --H-- needs chocolate. - From owner-freebsd-scsi Sun Mar 9 08:10:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA26727 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 08:10:34 -0800 (PST) Received: from zibbi.mikom.csir.co.za (zibbi.mikom.csir.co.za [146.64.24.58]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA26720 for ; Sun, 9 Mar 1997 08:10:30 -0800 (PST) Received: (from jhay@localhost) by zibbi.mikom.csir.co.za (8.8.5/8.8.5) id SAA25371; Sun, 9 Mar 1997 18:10:16 +0200 (SAT) From: John Hay Message-Id: <199703091610.SAA25371@zibbi.mikom.csir.co.za> Subject: Re: Micropolis 4221W and 20MB/s SYNC In-Reply-To: <199703091429.JAA00689@weenix.guru.org> from Keith Mitchell at "Mar 9, 97 09:29:07 am" To: kmitch@weenix.guru.org (Keith Mitchell) Date: Sun, 9 Mar 1997 18:10:16 +0200 (SAT) Cc: scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > For some reason my Micropolis 4221W (wide scsi) drive is only getting > 10 MB/s sync rate instead of the 20 it is supposed to get (according to > the drive spec). The drive is hanging off an Adaptec 3940UW controller. > > The sync rate in the setup is set to 20 and ultra is disabled (enabling it > doesn't do anything). > > I am running 2.2-GAMMA from 3/05. Below is the verbose probe message for > the drive. > > ahc0: target 0 using 16Bit transfers > ahc0: target 0 synchronous at 10.0MHz, offset = 0x8 > ahc0: target 0 Tagged Queuing Device > (ahc0:0:0): "MICROP 4221-09 1128RF 28RF" type 0 fixed SCSI 2 > sd is configured at 0 > sd0(ahc0:0:0): Direct-Access 1955MB (4004219 512 byte sectors) > sd0(ahc0:0:0): with 4048 cyls, 9 heads, and an average 109 sectors/track > > As you can see it uses wide transfers. It just is operating at hakf the > speed it should. > Are you sure? The way I read it, is 16Bit (2 bytes) and 10MHz which gives 20MB/s. John -- John Hay -- John.Hay@mikom.csir.co.za From owner-freebsd-scsi Sun Mar 9 14:28:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA16090 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 14:28:11 -0800 (PST) Received: from verdi.nethelp.no (verdi.nethelp.no [193.91.212.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA16074 for ; Sun, 9 Mar 1997 14:28:05 -0800 (PST) From: sthaug@nethelp.no Received: (qmail 4060 invoked by uid 1001); 9 Mar 1997 22:27:40 +0000 (GMT) To: freebsd-scsi@freebsd.org Subject: How to check synch transfer rate; can 2940UW do wide transfers with narrow device attached? X-Mailer: Mew version 1.05+ on Emacs 19.28.2 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sun, 09 Mar 1997 23:27:40 +0100 Message-ID: <4058.857946460@verdi.nethelp.no> Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have a machine running 2.2-970215-GAMMA with an Adaptec 2940UW, which is identified by the system as: ahc0 rev 0 int a irq 10 on pci0:16 ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs Two questions concerning this: 1. Assuming a wide external device (in my case a Seagate ST15150W) *and* a narrow internal device (IBM Ultrastar 2ES) connected at the same time: Can the Adaptec still do wide transfers against the wide disk, or are all transfers restricted to narrow? 2. How can I check what transfer rate is actually used for a particular device? Steinar Haug, Nethelp consulting, sthaug@nethelp.no From owner-freebsd-scsi Sun Mar 9 15:40:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA20705 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 15:40:39 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA20700 for ; Sun, 9 Mar 1997 15:40:36 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id PAA14163; Sun, 9 Mar 1997 15:41:12 -0800 (PST) Message-Id: <199703092341.PAA14163@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Keith Mitchell cc: scsi@freebsd.org Subject: Re: Micropolis 4221W and 20MB/s SYNC In-reply-to: Your message of "Sun, 09 Mar 1997 09:29:07 EST." <199703091429.JAA00689@weenix.guru.org> From: David Greenman Reply-To: dg@root.com Date: Sun, 09 Mar 1997 15:41:12 -0800 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >For some reason my Micropolis 4221W (wide scsi) drive is only getting >10 MB/s sync rate instead of the 20 it is supposed to get (according to >the drive spec). The drive is hanging off an Adaptec 3940UW controller. > >The sync rate in the setup is set to 20 and ultra is disabled (enabling it >doesn't do anything). Ultra mode has to be enabled in SCSI-select in order to use Ultra speeds. The problem is that the Micropolis 4221 isn't an Ultra-SCSI drive. The product literature clearly says "fast/wide" which means 10MHz, 16bits...or in other words, 20MBytes/sec. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-scsi Sun Mar 9 15:49:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA20969 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 15:49:25 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA20964 for ; Sun, 9 Mar 1997 15:49:23 -0800 (PST) Received: from narnia.plutotech.com (localhost [127.0.0.1]) by narnia.plutotech.com (8.8.5/8.7.3) with ESMTP id QAA06141; Sun, 9 Mar 1997 16:49:25 -0700 (MST) Message-Id: <199703092349.QAA06141@narnia.plutotech.com> To: sthaug@nethelp.no cc: freebsd-scsi@FreeBSD.ORG Subject: Re: How to check synch transfer rate; can 2940UW do wide transfers with narrow device attached? In-reply-to: Your message of "Sun, 09 Mar 1997 23:27:40 +0100." <4058.857946460@verdi.nethelp.no> Date: Sun, 09 Mar 1997 16:49:25 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Two questions concerning this: > >1. Assuming a wide external device (in my case a Seagate ST15150W) *and* >a narrow internal device (IBM Ultrastar 2ES) connected at the same time: >Can the Adaptec still do wide transfers against the wide disk, or are all >transfers restricted to narrow? Each device will be handled to the maximum advantage of its ability. >2. How can I check what transfer rate is actually used for a particular >device? Boot with the '-v' flag. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-scsi Sun Mar 9 23:11:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA11520 for freebsd-scsi-outgoing; Sun, 9 Mar 1997 23:11:25 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA11449; Sun, 9 Mar 1997 23:11:07 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA04038; Mon, 10 Mar 1997 00:24:40 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id AAA29409; Mon, 10 Mar 1997 00:07:18 +0100 (MET) Message-ID: <19970310000718.MT14198@uriah.heep.sax.de> Date: Mon, 10 Mar 1997 00:07:18 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: steve@vic.cioe.com (Steve Ames) Cc: freebsd-questions@FreeBSD.ORG, freebsd-scsi@FreeBSD.ORG Subject: Re: Conner tapestor problems? References: <199703071751.MAA20405@vic.cioe.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199703071751.MAA20405@vic.cioe.com>; from Steve Ames on Mar 7, 1997 12:51:41 -0500 Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Steve Ames wrote: > I have a conner tapestor (TSM4000R) SCSI tape backup. Thats one of them > QIC-3080 jobs. I'm using Sony QW3080XLF tapes (what the drive recommended). > > Problem seems to be that I'm only getting use about an eighth of the tape. > I run the backup of my system and get an I/O error (or a screenful of > MEDIUM errors). That means you've got -- well, medium errors. I'm not very familar with this wannabe-QIC technology. Maybe it's possible to reformat the cartridges? Try scsi -f /dev/st0.ctl -s 7200 -c "4 0 0 0 0 0" (Btw., which kind of medium errors do you get?) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Mon Mar 10 00:58:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA04683 for freebsd-scsi-outgoing; Mon, 10 Mar 1997 00:58:47 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA04678 for ; Mon, 10 Mar 1997 00:58:41 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA13512; Mon, 10 Mar 1997 09:58:24 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id JAA08143; Mon, 10 Mar 1997 09:55:57 +0100 (MET) Message-ID: <19970310095557.GH50073@uriah.heep.sax.de> Date: Mon, 10 Mar 1997 09:55:57 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: gilbertd@cs.man.ac.uk (Dave Gilbert) Cc: scsi@freebsd.org Subject: Re: Medium error messages - how to read References: <3322D5EA.9398D0B@cs.man.ac.uk> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3322D5EA.9398D0B@cs.man.ac.uk>; from Dave Gilbert on Mar 9, 1997 15:23:22 +0000 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Dave Gilbert wrote: > Mar 8 21:57:40 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb ^^^^^ That's the failing block number (in hex). > asc:11,0 U > nrecovered read error > Now I'm presuming that the 2940UW driver is not lying and it really > is a Medium error. The driver can hardly lye about this. > First of all what should I do - is there anyway I can see if the drive > has automatic block remapping on? Yes, it's on mode page 1. Run scsi -f /dev/rsd1.ctl -m 1 -e ...and set AWRE and AWWE to 1. Anyway, since it's an unrecovered error, you still need to write the block once in order to force a remap. > I'm inclined to think its the drive; but the fact that its only > happened after its been put on the PC makes me suspicious - even > though it is the 3rd one of these drives that we've had! I think the Sun handled remapping of its own. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Mon Mar 10 01:47:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA07065 for freebsd-scsi-outgoing; Mon, 10 Mar 1997 01:47:19 -0800 (PST) Received: from verdi.nethelp.no (verdi.nethelp.no [193.91.212.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA07057 for ; Mon, 10 Mar 1997 01:47:14 -0800 (PST) From: sthaug@nethelp.no Received: (qmail 6273 invoked by uid 1001); 10 Mar 1997 09:47:02 +0000 (GMT) To: steve@vic.cioe.com Cc: joerg_wunsch@uriah.heep.sax.de, freebsd-questions@freebsd.org, freebsd-scsi@freebsd.org Subject: Re: Conner tapestor problems? In-Reply-To: Your message of "Mon, 10 Mar 1997 00:07:18 +0100" References: <19970310000718.MT14198@uriah.heep.sax.de> X-Mailer: Mew version 1.05+ on Emacs 19.28.2 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Mon, 10 Mar 1997 10:47:02 +0100 Message-ID: <6271.857987222@verdi.nethelp.no> Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I have a conner tapestor (TSM4000R) SCSI tape backup. Thats one of them > > QIC-3080 jobs. I'm using Sony QW3080XLF tapes (what the drive recommended). > > > > Problem seems to be that I'm only getting use about an eighth of the tape. > > I run the backup of my system and get an I/O error (or a screenful of > > MEDIUM errors). > > That means you've got -- well, medium errors. > > I'm not very familar with this wannabe-QIC technology. Maybe it's > possible to reformat the cartridges? Try > > scsi -f /dev/st0.ctl -s 7200 -c "4 0 0 0 0 0" > > (Btw., which kind of medium errors do you get?) I have one of these, and it's working just fine for me. I use the same tapes, and I get very close to 2 GB (uncompressed) per tape. Steinar Haug, Nethelp consulting, sthaug@nethelp.no From owner-freebsd-scsi Tue Mar 11 16:15:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA22316 for freebsd-scsi-outgoing; Tue, 11 Mar 1997 16:15:08 -0800 (PST) Received: from caliban.dihelix.com (caliban.mrtc.org [199.4.33.251]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA22306 for ; Tue, 11 Mar 1997 16:15:04 -0800 (PST) Received: (from langfod@localhost) by caliban.dihelix.com (8.8.4/8.8.3) id OAA07716; Tue, 11 Mar 1997 14:15:01 -1000 (HST) Date: Tue, 11 Mar 1997 14:15:01 -1000 (HST) From: David Langford Message-Id: <199703120015.OAA07716@caliban.dihelix.com> To: asami@vader.cs.berkeley.edu, scsi@freebsd.org Subject: CCD and "spindle sync" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Would it help or be advantageous to try to find the "spindle sync" cable on the drives in a CCD array? I am putting together a small system with two 3940's and four Seagate Barrcuda's (new working ones I hope) was curious about this. Thanks, David Langford langfod@dihelix.com From owner-freebsd-scsi Thu Mar 13 06:51:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA00662 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 06:51:47 -0800 (PST) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA00657 for ; Thu, 13 Mar 1997 06:51:39 -0800 (PST) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.8.5/8.8.3) id QAA18460; Thu, 13 Mar 1997 16:48:51 +0200 (EET) From: mika ruohotie Message-Id: <199703131448.QAA18460@shadows.aeon.net> Subject: Re: Still SCSI Bus resets with 2940 on FreeBSD-2.2 In-Reply-To: <19970307195710.08931@klemm.gtn.com> from Andreas Klemm at "Mar 7, 97 07:57:10 pm" To: andreas@klemm.gtn.com (Andreas Klemm) Date: Thu, 13 Mar 1997 16:48:51 +0200 (EET) Cc: scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Sorry to say this, but I still have serious trouble with my AHA > 2940 when dumping data to my QIC Streamer (TDC-4222) and tagged > command queuing is enabled. > > Ordered Tag queued > sd0(ahc0:0:0): SCB 0x4 - timed out in command phase, SCSISIGI==0xc6 > SEQADDR==0x5e > sd0(ahc0:0:0): abort message in message buffer > sd0(ahc0:0:0): SCB7 - Abort Completed > sd0(ahc0:0:0): no longer in timeout > Ordered Tag sent this might sound weird, but _these_ were the messages i got with my 2.1.STABLE machine (originally ccded as 2.1.5 then went to 2.1.7), without configuring > options AHC_TAGENABLE > options AHC_SCBPAGING_ENABLE > options AHC_ALLOW_MEMIO these in the kernel. i never have gotten these on any kernel so far. and _yes_, the machine did hang too... i had screen fulls of those... but since it was a serious panic nothing got logged... i dont dump panics... and the question is, if those should be visible when AHC_* options are configured into the kernel, why did i see them? for people who dont remember or havent seen my earlier articles about this, i used ccd, but after i saw those i got scared and took them "down". (ccd with and without mirror, interleaved and not) (i have posted few articles about this to the hackers too, but i think i dont have to crosspost this to there) mickey From owner-freebsd-scsi Thu Mar 13 07:05:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA01281 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 07:05:15 -0800 (PST) Received: from shadows.aeon.net (shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA01272 for ; Thu, 13 Mar 1997 07:05:08 -0800 (PST) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.8.5/8.8.3) id RAA18593; Thu, 13 Mar 1997 17:01:47 +0200 (EET) From: mika ruohotie Message-Id: <199703131501.RAA18593@shadows.aeon.net> Subject: Re: Medium error messages - how to read In-Reply-To: <3322D5EA.9398D0B@cs.man.ac.uk> from Dave Gilbert at "Mar 9, 97 03:23:22 pm" To: gilbertd@cs.man.ac.uk (Dave Gilbert) Date: Thu, 13 Mar 1997 17:01:47 +0200 (EET) Cc: scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Mar 8 21:57:40 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb > asc:11,0 U > nrecovered read error > Mar 8 21:57:40 uriah /kernel: , retries:4 > > Now I'm presuming that the 2940UW driver is not lying and it really > is a Medium error. ugh, i dont know where to trust anymore... my fujitsu drives gave those to me in the "early" days, and in the last ones as well, but with those it all started... 3940UW was the adapter... (i have posted the exact specs atleast several times) as i just posted on the other article, i saw SCB stuff without enabling them after a while (with ccd)... i hope you are not running ccd system, are you? i got my data probably mostly out, didnt even bother trying the /usr/src (filesystem that seemed to be the one most fucked up) or other irrelevant filesystems... atleast once i got away from those MEDIUM ERRORs by newfs:ing, i was able to cpio tyhings out from the fs. (except in the end when the SCB things came and froze it when i attempted cpio) funny thing in my ERRORs were the fact that i occasionally got them and other SCSI anomalities when running, but not when rebooting, on the reboot mostly the fsck did finish nicely, while running it didnt, in the end it did finish only if i ran it manually, not in the boot. so there's something lurking into the ahc, right? mickey From owner-freebsd-scsi Thu Mar 13 07:16:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA01983 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 07:16:13 -0800 (PST) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA01976 for ; Thu, 13 Mar 1997 07:16:08 -0800 (PST) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.8.5/8.8.3) id RAA18662; Thu, 13 Mar 1997 17:12:30 +0200 (EET) From: mika ruohotie Message-Id: <199703131512.RAA18662@shadows.aeon.net> Subject: Re: Medium error messages - how to read In-Reply-To: <19970310095557.GH50073@uriah.heep.sax.de> from J Wunsch at "Mar 10, 97 09:55:57 am" To: joerg_wunsch@uriah.heep.sax.de Date: Thu, 13 Mar 1997 17:12:30 +0200 (EET) Cc: gilbertd@cs.man.ac.uk, scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > First of all what should I do - is there anyway I can see if the drive > > has automatic block remapping on? > Yes, it's on mode page 1. Run > scsi -f /dev/rsd1.ctl -m 1 -e > ...and set AWRE and AWWE to 1. Anyway, since it's an unrecovered > error, you still need to write the block once in order to force a > remap. hmm... since on my fujitsu drives these were both set to 0, does it mean that whatever problems i had were coz of it and not coz of ahc, coz suddenly it sounds like that to me? *blush* it it's true, once again it was proved that getting cheapo hardware is not good trade off... not that _i_ were recommending purchasing those drives... i set them to 1, btw. thanx, let's see what happens now. =) ofcourse i should learn to doubt the hw first, freebsd second. i should know, i should know, i should know. i'll install two more freebsd servers coz of this. *sulk* mickey From owner-freebsd-scsi Thu Mar 13 07:39:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03083 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 07:39:05 -0800 (PST) Received: from m1.cs.man.ac.uk (0@m1.cs.man.ac.uk [130.88.13.4]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA03076 for ; Thu, 13 Mar 1997 07:39:02 -0800 (PST) Received: from amu7.cs.man.ac.uk by m1.cs.man.ac.uk (4.1/SMI-4.1:AL6) id AA28691; Thu, 13 Mar 97 15:38:54 GMT Date: Thu, 13 Mar 97 15:38:54 GMT From: David Alan Gilbert Message-Id: <9703131538.AA26393@amu7.cs.man.ac.uk> To: scsi@freebsd.org Subject: Re: Medium errors Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Well I flipped AWRE etc on - Previously where dd'ing reading the disc would give log messages followed by an I/O error in the DD, now it managed to kill the machine and cause it to reboot itself. Great. The drive surface checked on a Sun; so I scsiformat'd it back on the PC and it seems OK. However: 1) Can someone explain to me what the Sun is doing with defect lists? How do I look at the 'manufacturers defect list' on FreeBSD - and what does it affect? 2) The disklabel stuff STILL doesn't work. I tried zero'ing the disc first; but running disklabel gives me various IOCTL errors. I tried using the /stand/sysinstall program and it happily fdisk'd but then segfaulted when it tried to label it. In the end I dd'd the partition and disklabel information from another (different sized disc) and then edited it - once a sensible partition/disklabel is on its happy. I think it must be a kernel related thing because the version on the boot disc works OK. Dave From owner-freebsd-scsi Thu Mar 13 07:41:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03218 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 07:41:22 -0800 (PST) Received: from sand.sentex.ca (sand.sentex.ca [206.222.77.6]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA03206 for ; Thu, 13 Mar 1997 07:41:18 -0800 (PST) Received: from gravel (gravel.sentex.ca [205.211.165.210]) by sand.sentex.ca (8.8.5/8.8.3) with SMTP id KAA00363; Thu, 13 Mar 1997 10:43:34 -0500 (EST) Message-Id: <3.0.1.32.19970313102806.00b212c0@sentex.net> X-Sender: mdtancsa@sentex.net X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Thu, 13 Mar 1997 10:28:06 -0500 To: mika ruohotie , gilbertd@cs.man.ac.uk (Dave Gilbert) From: Mike Tancsa Subject: Re: Medium error messages - how to read Cc: scsi@freebsd.org In-Reply-To: <199703131501.RAA18593@shadows.aeon.net> References: <3322D5EA.9398D0B@cs.man.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 05:01 PM 3/13/97 +0200, mika ruohotie wrote: >> Mar 8 21:57:40 uriah /kernel: sd1(ahc0:2:0): MEDIUM ERROR info:756cb >> asc:11,0 U >> nrecovered read error >> Mar 8 21:57:40 uriah /kernel: , retries:4 >> >> Now I'm presuming that the 2940UW driver is not lying and it really >> is a Medium error. > >ugh, i dont know where to trust anymore... my fujitsu drives gave those Interesting that you should mention Fujitsu... I have a couple of the 4gig'rs, and I found that when I created a better cooling system for the drives (I put a CPU fan on the front of my tower case to provide better circulation), I reduced a lot of these errors... Perhaps the drive is running too hot ? Its worth a try.. It helped my situation. ---Mike ********************************************************************** Mike Tancsa (mike@sentex.net) * To do is to be -- Nietzsche Sentex Communications Corp, * To be is to do -- Sartre Cambridge, Ontario * Do be do be do -- Sinatra (http://www.sentex.net/~mdtancsa) * From owner-freebsd-scsi Thu Mar 13 07:47:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03558 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 07:47:15 -0800 (PST) Received: from hda.hda.com (ip66-max1-fitch.ziplink.net [199.232.245.66]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA03553 for ; Thu, 13 Mar 1997 07:47:11 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.6.12/8.6.12) id JAA07232; Thu, 13 Mar 1997 09:03:43 -0500 From: Peter Dufault Message-Id: <199703131403.JAA07232@hda.hda.com> Subject: Re: Medium error messages - how to read In-Reply-To: <199703131512.RAA18662@shadows.aeon.net> from mika ruohotie at "Mar 13, 97 05:12:30 pm" To: bsdscsi@shadows.aeon.net (mika ruohotie) Date: Thu, 13 Mar 1997 09:03:42 -0500 (EST) Cc: joerg_wunsch@uriah.heep.sax.de, gilbertd@cs.man.ac.uk, scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > First of all what should I do - is there anyway I can see if the drive > > > has automatic block remapping on? > > Yes, it's on mode page 1. Run > > scsi -f /dev/rsd1.ctl -m 1 -e > > ...and set AWRE and AWWE to 1. Anyway, since it's an unrecovered > > error, you still need to write the block once in order to force a > > remap. > > hmm... since on my fujitsu drives these were both set to 0, does it mean > that whatever problems i had were coz of it and not coz of ahc, coz suddenly > it sounds like that to me? *blush* > > it it's true, once again it was proved that getting cheapo hardware is not > good trade off... not that _i_ were recommending purchasing those drives... There's nothing wrong with the Fujitsu drives. I have eight of them in various places without any problems. You need to store the settings permanently by saving them in the non-volatile page or they will reset at the next power cycle. Use a "-P 3" with the same command. Here are a few other things to check (from zones.wish in scsinew in freefall's ~ftp/pub/dufault/scsinew.tgz) > # Mode Field Bad > # page value > set check(awre) {1 0 0 "Automatic write reallocation is not enabled"} > set check(arre) {1 1 0 "Automatic read reallocation is not enabled"} > set check(tb) {1 2 1 \ > "Data is transferred even if in error (but an error is generated)"} > set check(rc) {1 3 1 \ > "Data is transferred without delays even in the case of error"} > set check(eer) {1 4 1 \ > "The most expedient form of error recovery is used"} > set check(dcr) {1 7 1 "All error detection is disabled"} > set check(rrc) {1 8 0 "The read retry count is 0"} > set check(wrc) {1 12 0 "The write retry count is 0"} > set check(wce) {8 1 0 "The write cache is not enabled"} > set check(rcd) {8 3 1 "The read cache is disabled"} Peter -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 From owner-freebsd-scsi Thu Mar 13 13:58:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA23270 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 13:58:57 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA23265 for ; Thu, 13 Mar 1997 13:58:45 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA13446; Thu, 13 Mar 1997 22:56:46 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA00441; Thu, 13 Mar 1997 22:46:39 +0100 (MET) Message-ID: <19970313224639.RG19743@uriah.heep.sax.de> Date: Thu, 13 Mar 1997 22:46:39 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: gilbertd@cs.man.ac.uk (David Alan Gilbert) Cc: scsi@FreeBSD.ORG Subject: Re: Medium errors References: <9703131538.AA26393@amu7.cs.man.ac.uk> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <9703131538.AA26393@amu7.cs.man.ac.uk>; from David Alan Gilbert on Mar 13, 1997 15:38:54 +0000 Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As David Alan Gilbert wrote: > 1) Can someone explain to me what the Sun is doing with defect > lists? How do I look at the 'manufacturers defect list' on > FreeBSD - and what does it affect? There's a script now, in src/tools/tools/scsi-defects/scsi-defects.pl, courtesy Peter Dufault. Be warned, it hung my Fujitsu drive at work, but seems to run well on a variety of drives. > 2) The disklabel stuff STILL doesn't work. I tried zero'ing the disc > first; but running disklabel gives me various IOCTL errors. Define `various'. Explain exactly what you did. Using or not using -r makes a great difference on a pristine disk. Using your own idea of a device node, too (simply use `sd0' or such). This is what is supposed to work: dd if=/dev/zero of=/dev/rsd0 count=15 disklabel -Brw sd0 auto -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Thu Mar 13 17:41:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA04752 for freebsd-scsi-outgoing; Thu, 13 Mar 1997 17:41:07 -0800 (PST) Received: from pop.cybernex.net (root@pop.cybernex.net [204.141.116.15]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA04743 for ; Thu, 13 Mar 1997 17:41:02 -0800 (PST) Received: from skippy.grunfelder.com (nj009a-192.cybernex.net [204.141.117.192]) by pop.cybernex.net (8.8.5/8.6.12) with SMTP id UAA15392 for ; Thu, 13 Mar 1997 20:41:05 -0500 Message-Id: <3.0.1.32.19970313204041.007e5480@mail.cybernex.net> X-Sender: wjgrun@mail.cybernex.net X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Thu, 13 Mar 1997 20:40:41 -0500 To: freebsd-scsi@freebsd.org From: Bill Grunfelder Subject: Hardware Failure error message Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Any hints as to what this error message means? The drive seems to be running OK, but I get this error message 2 or 3 times a day. (And I don't like the way it sounds Mar 13 11:18:13 machinename /kernel: sd1(bt0:5:0): HARDWARE FAILURE asc:1,0 No index/sector signal Bill .......................................................................... Bill Grunfelder wjgrun@poboxes.com ................... http://planetx.bloomu.edu/~wjgrun/ ................... From owner-freebsd-scsi Fri Mar 14 00:32:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA07165 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 00:32:24 -0800 (PST) Received: from shadows.aeon.net (bsdscsi@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA07158 for ; Fri, 14 Mar 1997 00:32:19 -0800 (PST) Received: (from bsdscsi@localhost) by shadows.aeon.net (8.8.5/8.8.3) id KAA26719; Fri, 14 Mar 1997 10:28:48 +0200 (EET) From: mika ruohotie Message-Id: <199703140828.KAA26719@shadows.aeon.net> Subject: Re: Medium error messages - how to read In-Reply-To: <199703131403.JAA07232@hda.hda.com> from Peter Dufault at "Mar 13, 97 09:03:42 am" To: dufault@hda.com (Peter Dufault) Date: Fri, 14 Mar 1997 10:28:48 +0200 (EET) Cc: joerg_wunsch@uriah.heep.sax.de, gilbertd@cs.man.ac.uk, scsi@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [my scsi probs] > You need to store the settings permanently by saving them in the non-volatile > page or they will reset at the next power cycle. Use a "-P 3" with > the same command. why my drives disallow me to set the ARRE to 1? i tried multiple times. even if i edit that, when i look the value after save&exit it's 0 again. this is what i have now: pluto# scsi -f /dev/rsd1.ctl -m 1 AWRE (Auto Write Reallocation Enbld): 1 ARRE (Auto Read Reallocation Enbld): 0 TB (Transfer Block): 1 RC (Read Continuous): 0 EER (Enable Early Recovery): 0 PER (Post Error): 0 DTE (Disable Transfer on Error): 0 DCR (Disable Correction): 0 Read Retry Count: 34 Correction Span: 41 Head Offset Count: 0 Data Strobe Offset Count: 0 Write Retry Count: 34 Recovery Time Limit: 0 pluto# > > set check(rrc) {1 8 0 "The read retry count is 0"} > > set check(wrc) {1 12 0 "The write retry count is 0"} > > set check(wce) {8 1 0 "The write cache is not enabled"} > > set check(rcd) {8 3 1 "The read cache is disabled"} those werent listed, relevant? actually i looked other pages (without knowing actually what i was doing except that i assumed it being "safe") pluto# scsi -f /dev/rsd1.ctl -m 8 WCE: 0 MF: 0 RCD: 0 Demand Retention Priority: 0 Write Retention Priority: 0 Disable Pre-fetch Transfer Length: 65535 Minumum Pre-fetch: 0 Maximum Pre-fetch: 249 Maximum Pre-fetch Ceiling: 65535 should i edit that wce value to 1? (well, i did try, and it returns 0 just like the other value i twiddled with) (for the record on my quantum drive those all were "right" as default) pluto# scsi -f /dev/rsd1.ctl -m 3 Tracks per Zone: 5 Alternate Sectors per Zone: 11 Alternate Tracks per Zone: 0 Alternate Tracks per Logical Unit: 5 Sectors per Track: 186 Data Bytes per Physical Sector: 512 Interleave: 1 Track Skew Factor: 36 Cylinder Skew Factor: 36 SSEC: 0 HSEC: 1 RMB: 0 SURF: 0 pluto# scsi -f /dev/rsd1.ctl -m 4 Number of Cylinders: 5714 Number of Heads: 5 Starting Cylinder-Write Precompensation: 0 Starting Cylinder-Reduced Write Current: 0 Drive Step Rate: 0 Landing Zone Cylinder: 0 RPL: 0 Rotational Offset: 0 Medium Rotation Rate: 7200 pluto# scsi -f /dev/rsd1.ctl -m 0 96 03 00 00 00 00 00 00 00 00 00 pluto# scsi -f /dev/rsd1.ctl -m 2 Buffer Full Ratio: 128 Buffer Empty Ratio: 128 Bus Inactivity Limit: 50 Disconnect Time Limit: 0 Connect Time Limit: 0 Maximum Burst Size: 0 DTDC: 0 does that look sane? as it seems obvious, i only use scsi devices. > Peter mickey From owner-freebsd-scsi Fri Mar 14 00:41:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA07523 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 00:41:22 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA07516 for ; Fri, 14 Mar 1997 00:41:10 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA23773; Fri, 14 Mar 1997 09:41:04 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id JAA22756; Fri, 14 Mar 1997 09:17:29 +0100 (MET) Message-ID: <19970314091726.QU31302@uriah.heep.sax.de> Date: Fri, 14 Mar 1997 09:17:26 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: wjgrun@poboxes.com (Bill Grunfelder) Cc: freebsd-scsi@freebsd.org Subject: Re: Hardware Failure error message References: <3.0.1.32.19970313204041.007e5480@mail.cybernex.net> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.1.32.19970313204041.007e5480@mail.cybernex.net>; from Bill Grunfelder on Mar 13, 1997 20:40:41 -0500 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Bill Grunfelder wrote: > Any hints as to what this error message means? With two or three people per week asking this question, seems we ought to right a FAQ entry for it... :) > The drive seems to be running OK, but I get this error message 2 or > 3 times a day. (And I don't like the way it sounds I wouldn't like it either. > Mar 13 11:18:13 machinename /kernel: sd1(bt0:5:0): HARDWARE FAILURE asc:1,0 > No index/sector signal I think this is pretty obvious, isn't it? The drive doesn't find some of its synchronization marks. I wouldn't trust it for any important data. Perhaps reformatting might help you. /sbin/scsiformat -wq sd1... I've got a drive that sometimes gets a lot of them: /kernel: sd0(aha0:0:0): MEDIUM ERROR info:1e030 asc:14,0 Recorded entity not found field replaceable unit: 38 /kernel: , retries:3 /kernel: sd0(aha0:0:0): MEDIUM ERROR info:1e030 asc:14,0 Recorded entity not found field replaceable unit: 38 /kernel: , retries:2 /kernel: sd0(aha0:0:0): MEDIUM ERROR info:1e030 asc:14,0 Recorded entity not found field replaceable unit: 38 /kernel: , retries:1 Anyway, it's 1.2 GB of scratch space, and despite of these messages, i have even seen `make world's completing on this drive. So i don't throw it away but use it for experimenting, release tests, SCSI device driver stress tests :) etc. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Fri Mar 14 01:12:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA08880 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 01:12:44 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA08854 for ; Fri, 14 Mar 1997 01:12:38 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id KAA24360 for scsi@freebsd.org; Fri, 14 Mar 1997 10:12:36 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id KAA29769; Fri, 14 Mar 1997 10:09:36 +0100 (MET) Message-ID: <19970314100936.DL62124@uriah.heep.sax.de> Date: Fri, 14 Mar 1997 10:09:36 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: scsi@freebsd.org Subject: Re: Medium error messages - how to read References: <199703131403.JAA07232@hda.hda.com> <199703140828.KAA26719@shadows.aeon.net> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199703140828.KAA26719@shadows.aeon.net>; from mika ruohotie on Mar 14, 1997 10:28:48 +0200 Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As mika ruohotie wrote: > why my drives disallow me to set the ARRE to 1? i tried multiple times. That probably means they don't support this feature at all. It's optional by the standard. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-scsi Fri Mar 14 01:54:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA10827 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 01:54:05 -0800 (PST) Received: from hda.hda.com (ip89-max1-fitch.ziplink.net [199.232.245.89]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA10822 for ; Fri, 14 Mar 1997 01:54:02 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.6.12/8.6.12) id DAA09247; Fri, 14 Mar 1997 03:10:43 -0500 From: Peter Dufault Message-Id: <199703140810.DAA09247@hda.hda.com> Subject: Re: Hardware Failure error message In-Reply-To: <19970314091726.QU31302@uriah.heep.sax.de> from J Wunsch at "Mar 14, 97 09:17:26 am" To: joerg_wunsch@uriah.heep.sax.de Date: Fri, 14 Mar 1997 03:10:43 -0500 (EST) Cc: wjgrun@poboxes.com, freebsd-scsi@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I've got a drive that sometimes gets a lot of them: > > /kernel: sd0(aha0:0:0): MEDIUM ERROR info:1e030 asc:14,0 Recorded entity not found field replaceable unit: 38 > /kernel: , retries:3 You can supposedly get back the ascii text for the field replaceable units on some drives. If this isn't a Fujitsu (I've tried on two of them) I'll send you a script to see if you can read back what your FRU 38 is. Peter -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 From owner-freebsd-scsi Fri Mar 14 03:51:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA14443 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 03:51:47 -0800 (PST) Received: from m1.cs.man.ac.uk (0@m1.cs.man.ac.uk [130.88.13.4]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA14438 for ; Fri, 14 Mar 1997 03:51:41 -0800 (PST) Received: from amu7.cs.man.ac.uk by m1.cs.man.ac.uk (4.1/SMI-4.1:AL6) id AA20269; Fri, 14 Mar 97 11:51:17 GMT Date: Fri, 14 Mar 97 11:51:02 GMT From: David Alan Gilbert Message-Id: <9703141151.AA26882@amu7.cs.man.ac.uk> To: gilbertd@cs.man.ac.uk, joerg_wunsch@uriah.heep.sax.de Subject: Re: Medium errors Cc: scsi@FreeBSD.ORG Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > 1) Can someone explain to me what the Sun is doing with defect > > lists? How do I look at the 'manufacturers defect list' on > > FreeBSD - and what does it affect? > > There's a script now, in src/tools/tools/scsi-defects/scsi-defects.pl, > courtesy Peter Dufault. Be warned, it hung my Fujitsu drive at work, > but seems to run well on a variety of drives. Perhaps I didn't make myself clear; I want to understand more about the way defect lists are stored and processed. In Ye'olde MFM drive it was the OS which used the defect list while working on the drive; now I presume its the drives job? But if thats so why does the Sun have so much to firtle with it? Can you give me a full path, machine name and all to that - I can't find it. > > 2) The disklabel stuff STILL doesn't work. I tried zero'ing the disc > > first; but running disklabel gives me various IOCTL errors. > > Define `various'. Explain exactly what you did. Using or not using > -r makes a great difference on a pristine disk. Using your own idea > of a device node, too (simply use `sd0' or such). Sorry - I was in too much of a rush ; the server was own and I had colleagues screaming for their data :-) - didn't write the IOCTL name down. > This is what is supposed to work: > > dd if=/dev/zero of=/dev/rsd0 count=15 > disklabel -Brw sd0 auto No (well I didn't put the -B in); I fdisk'd first, but I also tried just fdisk'ing and just disklabeling (I think) - nothing gave a happy result. Dave From owner-freebsd-scsi Fri Mar 14 10:37:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA04417 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 10:37:26 -0800 (PST) Received: from vex.net (vex.net [207.207.191.193]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA04399; Fri, 14 Mar 1997 10:37:06 -0800 (PST) Received: from vex.net(really [207.207.191.193]) by vex.net via sendmail with smtp id for ; Fri, 14 Mar 1997 13:36:56 -0500 (EST) (Smail-3.2.0.90 1996-Dec-4 #4 built 1997-Jan-8) Date: Fri, 14 Mar 1997 13:36:56 -0500 (EST) From: Brian Tao X-Sender: taob@vex.net To: FREEBSD-SCSI-L , FREEBSD-HACKERS-L Subject: FreeBSD DPT RAID drivers Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Here's a quick update on the FreeBSD drivers for the DPT SCSI-to-SCSI RAID controllers for PC's. -- Brian Tao (BT300, taob@risc.org) "Though this be madness, yet there is method in't" ---------- Forwarded message ---------- Date: Fri, 7 Mar 1997 12:35:13 -0800 (PST) From: Michael Neuffer To: inet-access@earth.com Subject: Re: FreeBSD vs. Linux On Fri, 7 Mar 1997, Richard Stanford wrote: >What are they wanting to use this server for? Linux fans will point >out that it can handle all sorts of weird (read: not good for a mission- >critical server) hardware (although it /does/ have support for that DPT >RAID card, which FreeBSD doesn't (grumble (damn, wish I could write a >decent device driver...))). The next generaton of the eata-dma driver (which is the driver for the DPTs) is currently under development. We are actually developing it for FreeBSD first and I am going to port it back to Linux once it is stable. It is going to have a number of new features that the current (2.59b) Linux release not has. Mike Michael Neuffer i-Connect.Net, a Division of iConnect Corp. mike@i-Connect.Net 14355 SW Allen Blvd., Suite 140 503.641.8774 Beaverton, OR 97005 ============================== ISP Mailing List ============================== Email ``unsubscribe'' to inet-access-request@earth.com to be removed. Because this is a restricted list -- DO NOT CROSS POST TO OTHER LISTS. From owner-freebsd-scsi Fri Mar 14 13:23:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA12896 for freebsd-scsi-outgoing; Fri, 14 Mar 1997 13:23:53 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA12869 for ; Fri, 14 Mar 1997 13:23:19 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA11944; Fri, 14 Mar 1997 22:22:28 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA20101; Fri, 14 Mar 1997 22:01:02 +0100 (MET) Message-ID: <19970314220101.SM26356@uriah.heep.sax.de> Date: Fri, 14 Mar 1997 22:01:01 +0100 From: j@uriah.heep.sax.de (J Wunsch) To: scsi@FreeBSD.ORG Cc: gilbertd@cs.man.ac.uk (David Alan Gilbert) Subject: Re: Medium errors References: <9703141151.AA26882@amu7.cs.man.ac.uk> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Sender: owner-freebsd-scsi@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As David Alan Gilbert wrote: > Perhaps I didn't make myself clear; I want to understand more about > the way defect lists are stored and processed. In Ye'olde MFM drive > it was the OS which used the defect list while working on the drive; > now I presume its the drives job? But if thats so why does the Sun > have so much to firtle with it? With SCSI, you can normally have both ways. Normally, the drive will handle its bad block lists, and provide automatic remapping. For most drives, you can also turn it off, and do the remapping in software. I've seen some commercial systems doing this, probably mostly since they could use the same code for ST-506 and SCSI by this, and get an error notification and consistent media defect handling between ST-506, ESDI, and SCSI. FreeBSD relies on the drive's remapping features, and the SCSI disk driver doesn't even offer you the opportunity to use bad144. > Can you give me a full path, machine name and all to that - I can't > find it. src/tools/tools/scsi-defects/scsi-defects.pl Relative to the source tree base of your closest FreeBSD mirror, in FreeBSD-current. (It's also in 2.2-GAMMA now, but only since a couple of days.) > > dd if=/dev/zero of=/dev/rsd0 count=15 > > disklabel -Brw sd0 auto > > No (well I didn't put the -B in); I fdisk'd first, but I also tried > just fdisk'ing and just disklabeling (I think) - nothing gave > a happy result. Probably. disklabel auto might not cooperate well with fdisk- organized disks, i never tested this case in any way, nor am i very interested in fdisk-organized disks at all. I simply consider the fdisk table something fairly obsoletish on my systems. (And have been cursing the HP folks since their newest NetServer LC even *ignores* its hard disk if the BIOS thinks it's smarter and the fdisk table doesn't look what it believes were reasonable! HP is a true Windows company these days.) The only thing i guarantee you for disklabel auto is that the above sequence is supposed to work. I don't promise you anything about fdisk (obviously :). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)